Contents

#!/usr/bin/env ruby

require 'open-uri'
require 'pstore'
require 'tempfile'

puts 'downloading...'

data = PStore.new('lib/polish_postal_codes.pstore')
tmp_file = Tempfile.new('spispna.pdf')
tmp_file.write(open('https://www.poczta-polska.pl/hermes/uploads/2013/11/spispna.pdf').read)
tmp_file.flush

puts 'converting...'

system("pdftotext -layout #{tmp_file.path}")

data.transaction do
  File.open("#{tmp_file.path}.txt", 'r').each_line do |line|
    break if line =~ /Podmioty posiadające odrębne PNA/
    next unless line =~ /([0-9]{2}-[0-9]{3})/

    t = line.gsub(/\ {2,}/, ' ').split(/ /)
    data[t[0].strip] ||= { city: t[1].split('(').first.strip,
                           community: t[-3].strip,
                           county: t[-2].strip,
                           province: t[-1].strip }
  end
end

at_exit do
  converted_txt_file_path = "#{tmp_file.path}.txt"
  File.delete converted_txt_file_path if File.exist? converted_txt_file_path
end

Version data entries

1 entries across 1 versions & 1 rubygems

Version	Path
polish_postal_codes-0.0.2	bin/update